OBJECTIVE: To evaluate the association of a standardized, structured approach to in-hospital postcesarean delivery pain management with maternal opioid use after cesarean delivery. METHODS: We conducted a retrospective cohort study of women who underwent cesarean delivery before and after a quality improvement intervention at a single tertiary care center. A multidisciplinary task force revised electronic order sets for all patients who underwent cesarean delivery with neuraxial anesthesia. The revised order set separated acetaminophen from opioids, scheduled acetaminophen and nonsteroidal antiinflammatory drug administration, and limited opioid use to breakthrough pain. Data were collected by electronic chart review. The primary outcome was median morphine milligram equivalents per hospital stay. Secondary outcomes included median morphine milligram equivalents per day, median pain scores, time to discharge, and opioid–nonopioid pain medication use. Descriptive and bivariable analyses were performed. RESULTS: There were no significant differences in baseline characteristics in the preintervention (n=283) and postintervention (n=286) groups. There was a 75% reduction in median morphine milligram equivalents per stay from 120 (90–176 interquartile range) preintervention to 30 (5–68) postintervention (P<.001) and a 77% reduction in median morphine milligram equivalents per day (51 [41–60] vs 12 [2–25], P<.001). There was no difference between groups in time to discharge or median pain scores. There was no difference in ketorolac use (80% preintervention vs 75% postintervention, P=.14) or in median ibuprofen mg per day (1,391 preintervention vs 1,347 postintervention, P=.22). There was an increase in median acetaminophen mg per day (753 preintervention vs 2,340 postintervention, P<.001). There was a significant increase in patients who used no opioids during their hospital stay (6% preintervention vs 19% postintervention, P<.001). CONCLUSION: A multimodal stepwise approach to postcesarean delivery pain control was associated with markedly reduced opioid consumption without increasing hospital stay or median pain scores. By separating acetaminophen from opioids and limiting opioids to breakthrough pain, we were able to operationalize a tier-based approach to pain management.

Objective To evaluate the ability of estimated blood loss (EBL) and quantitative blood loss (QBL) to predict the need for blood transfusion in postpartum patients. Methods This is a retrospective observational study involving all deliveries one year before and after the change from EBL to QBL assessment in June 2017. Blood loss, need for blood transfusion, admission hematocrit, and postpartum nadir hematocrit were collected. Descriptive and bivariable analyses were performed. Receiver operator curves were compared. Results Overall, the baseline characteristics between the EBL (n=2743) and QBL (n=2,712) groups were similar. Although there was a higher rate of blood loss ≥ 1,000 mL in QBL vs EBL (6.5% vs 2.1%, P<0.001), there was no difference in the rate of blood transfusions (2.0% vs 2.0%, P=1). Among cesarean deliveries, QBL outperformed EBL for predicting blood transfusion and/or ≥10 point drop in hematocrit (AUC 0.75 vs 0.66, P=0.02). QBL also outperformed EBL for predicting transfusion after vaginal delivery (AUC 0.93 vs 0.81, P=0.03). Conclusion QBL is a more sensitive test for detecting clinically significant blood loss, which could lead to earlier recognition of hemorrhage and interventions.

Diabetes is associated with increased risk of stillbirth and shoulder dystocia. Compared with uncomplicated pregnancies, diabetic patients have a 4-6x risk of stillbirth and 2-3x risk of shoulder dystocia. A 34 yo G2P0010 presented with a 40+3 wga IUFD with nonstandard antenatal glucose screening. Admission labs included a hemoglobin A1c of 6.6. She had a vaginal delivery complicated by a 30-minute shoulder dystocia that was not relieved by McRoberts, suprapubic pressure, Rubin II, Wood’s Screw, or posterior arm delivery. Nitroglycerine was administered, after which Wood’s Screw was successful resulting in delivery of an infant weighing 4190 grams (85th percentile for gestational age). A 31 yo G1 presented with a 37+1 wga IUFD. Her 28 wga three-hour GTT was notable for an elevated value at one hour (216 mg/dL). Admission labs included a hemoglobin A1c of 6.6. She had a vaginal delivery complicated by a 30-minute shoulder dystocia that was relieved via posterior axillary sling after failure of McRoberts, suprapubic pressure, Rubin II, Wood’s Screw, and Gaskin’s, resulting in the delivery of an infant weighing 3590 g (92nd percentile for gestational age). We present two cases of severe shoulder dystocia in patients who both presented with term IUFD and diabetic-range hemoglobin A1c. There is minimal literature on diabetic patients with pregnancies affected by both stillbirth and shoulder dystocia. These cases underscore the importance of glucose screening and control to prevent catastrophic obstetric outcomes.